Among human viruses etiologically linked to human cancer, the Epstein Barr virus (EBV) is unusual because of the diverse assortment of malignancies with which it is associated, if only sporadically. Our long term objective is to understand how a ubiquitous and persistent viral pathogen that normally achieves a lasting rapport with its host becomes pathogenic years after primary infection. Emerging possibilities that more fully acknowledge EBV opportunism in the context of the life-long carrier state include viral reactivation with entry into cells made more prone to infection and EBV-enhanced malignant progression by pre-existing molecular alterations. Here, we examine EBV tumorigenesis within a new conceptual framework of chance infection of neoplastic cells by virus endogenous to the host, with possible viral DNA loss from some tumors after EBV contribution to growth has been made. Aim 1 will use the pattern of EBV distribution and clonality in human tumor biopsies, as discerned by laser capture microdissection and single cell quantitative PCR, to establish a time frame for infection, subsequent clonal evolution, and EBV DNA loss. Aim 2 will determine how variably reiterated EBV terminal repeat (TR) sequences, comprising the first intron of the LMP2A gene, govern expression levels of this EBV oncoprotein. With TRs a commonly employed viral marker of tumor clonality, information obtained may implicate TR number per se in a process of cell selection that drives the transition from polyconal infection of a few tumor cells to monoclonal outgrowth of a subset with the highest LMP2A expression. Aim 3 will define the genomics of EBV redundancy and loss. In a newly devised in vitro system of transient infection, we will establish whether EBV transit through a cell produces epigenetic gene silencing as a component of the multistep process of carcinogenesis. The public health relevance of this research is that it will answer fundamental biological questions on the nature of EBV's erratic association with an ever expanding array of human tumors. Does EBV have the capability to initiate every tumor in which it is found as currently presupposed on the basis of EBV clonality or does it contribute instead to late stages of disease, entering cells more prone to infection and malignant progression by prior molecular mishaps? The information gained will further elucidate EBV's role in human cancer, its prognostic significance, and future therapeutic approaches.